How Automakers Can Survive the Self-Driving Era

Transcrição

How Automakers Can Survive the Self-Driving Era
How Automakers
Can Survive the
Self-Driving Era
A.T. Kearney study reveals new insights on who
will take the pole position in the $560 billion
autonomous driving race.
How Automakers Can Survive the Self-Driving Era
1
Executive Summary
Transportation as we know it would be impossible without the quantum leaps in technology
that have taken place over the past centuries. It is undeniable that the next big thing will
be autonomous driving. With this wave of innovation, traditional players in the automotive
industry could wind up in the passenger seat, with new entrants sitting pretty behind
the wheel.
For original equipment manufacturers to survive in this market, there are five key questions
to answer:
1. How can OEMs match consumer needs with autonomous driving solutions, while
overcoming skepticism about relinquishing control of the vehicle?
2. How will the market for autonomous driving develop and what will the associated product
roadmaps look like?
3. How will government legislation keep pace with new technologies while also addressing
questions of liability?
4. Which business models will win in the new industry?
5. What role will partner and competitor ecosystems play in autonomous driving?
To help answer these questions, A.T. Kearney conducted more than 150 interviews with executives at companies around the world who have a strong motivation to make autonomous driving
an affordable reality for consumers and for businesses. Our findings reveal crucial insights
along the five key questions:
Consumer needs
• The connected consumer prefers an individual lifestyle in a big city environment
• More urbanization results in intermodal mobility; services enabled by mobile devices provide
consumers with more flexibility and time
• Car ownership becomes less relevant than car-sharing services and platforms
Market and product roadmaps
• Changing consumer behavior causes a paradigm shift toward mobility as a service and
a preference for lavish private transportation
• The market for autonomous driving grows to $560 billion by 20351
• The main product categories around autonomous driving include mobile apps, special
equipment, autonomous cars, mobility services, and infrastructure
• Developed and mature markets, including Asian megacities, spearhead market development
and a global rollout
1
Market estimate is based on new car revenues, hardware upgrades, apps, and other digital features.
How Automakers Can Survive the Self-Driving Era
1
Legislation, technology, and liability
• Until 2025, legislation is the main roadblock to autonomous driving
• The most pressing legal issue is accident liability
• Achieving economic savings is the primary reason to drive legislation
Business models
• Autonomous driving threatens the very existence of mid-level automakers as the market
develops along three segments: premium, low-cost, and drones
• The industry splits in two—those that manufacture vehicles and those that provide
consumer services
• Revenues from pay-per-use services outperform optional equipment revenues from
2025 onward
Partner and competitor ecosystems
• Existing players in the automotive industry collaborate with new entrants to offer valueadded services
• Traditional OEMs have the first view on the consumer; the first OEM to build a value-added
service network with partners wins the market
In this report, we describe the forces shaping the autonomous driving market and what the
market will ultimately look like. We also define the core questions facing incumbent OEMs and
offer our recommendations on necessary preparations to compete in this market in terms of
product structures, business models, and teams.
How Automakers Can Survive the Self-Driving Era
2
A New Era of Mobility Is Upon Us
The next quantum leap in transportation is connected mobility. It combines the movement of
people, goods, and information into one elegant, consumer-friendly solution built on technology,
services, and stakeholders (see figure 1). Connected mobility—and the self-driving vehicles
that will serve it—puts major economies on the cusp of the first significant growth wave of the
21st century. This growth will be unprecedented, with an estimated annual value of around
$560 billion by 2035 for the core services around self-driving vehicles.
Self-driving vehicles, also referred to as autonomous driving, have a strong allure for consumers
and for companies within and outside the auto industry. These companies are fighting for the
pole position, jockeying to figure out how to compete and cooperate to serve consumers in this
vast market of seamless opportunities—from cross-channel consumer views and personalized
offerings to social media, in-store services, and cross-channel logistics.
Figure 1
Connected mobility is built on technology, services, and stakeholders
Technology
Smart grid
Smart
appliances
Carbon
management
Telecom
network
management
Multimodal
transport
Sustainability
control
Housing
Stakeholders
Residents
Telecoms
Energy
generation
management
Building
management
Services
One
consumerfriendly
solution
Visitors
Utilities
Information
Paperless
ticketing
Businesses
Identity
control
Smart
dashboard
Service
providers
Parking
Smart
data center
City
management
Entertainment
Carbon
monitoring
CRM
Electronic
ID
Security
Emergency
services
Sensors
(M2M)
Wayfinding
Geographic
information
systems
Content
management
Note: CRM is customer relationship management. M2M is machine to machine.
Sources: “The Digital Universe in 2020: Big Data, Bigger Digital Shadows, and Biggest Growth in the Far East” by EMC Corporation, Cisco Internet Business
Solutions Group; A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era
3
We recently conducted more than 150 interviews with industry executives worldwide, all of
whom are motivated to make autonomous driving an affordable reality for consumers and for
businesses. Their companies offer various levels of expertise. Some companies will create and
maintain the infrastructure and the vehicles. Others will capitalize on consumers’ newfound
freedom by offering a vast range of connected mobility and communications services.
As autonomous driving becomes a foregone conclusion, the most fascinating questions are
what will the market look like? How will driving patterns change in the decades ahead? How
will the economics of the massive automotive industry change as a result? We cannot answer
these questions without first determining who will win in this market, who will survive, and
who will lose. The answers will have dramatic and perhaps even fatal consequences for many
of today’s traditional car companies. The answers will also affect investors who must decide
where to place their bets.
In this report, we describe the forces shaping the autonomous driving market and define what
the market will ultimately look like. We outline the core questions facing incumbent OEMs and
offer our recommendations on necessary preparations to compete successfully in this market
in terms of consumers, products, and business models. We begin with the benefits to society.
The societal benefits of autonomous driving
Imagine a world with fewer accidents, fewer traffic deaths, greater energy efficiency, and lower
insurance premiums. This is the world of autonomous driving. It brings mobility to people who
lack easy or practical access to driving, such as the elderly and disabled. It also appeals to the
lifestyles and virtues embraced by millennials, such as health, entertainment, and mindfulness.
By the later stages of innovation, we, in accordance with the industry experts who participated
in our interviews, expect the following benefits to be especially powerful:
• Reduce traffic accidents by 70 percent, saving thousands of lives every year
• Lower vehicle service costs by 35 percent, in part because self-driving cars have far fewer
mechanical wear-and-tear parts
• Cut energy consumption by 30 percent, not only because of alternative energy sources but
also because automobiles can use car-to-car communications to swarm or travel in convoys
(platoon driving), thus improving efficiency and traffic flow
• Reduce insurance liability by more than 15 percent as driving becomes much safer and new
insurance models emerge
The estimated annual savings for the United States alone is expected to be around $1.3 trillion,
as shown in figure 2 on page 5.
Most drivers have already enjoyed the first wave of autonomous driving with features such as
navigation systems, in-car entertainment systems, lane-assist technologies, traffic warning
systems and sensors, and self-parking cars. In the second wave, Google, Tesla, and Uber are
aggressively popularizing the idea of self-driving cars (see sidebar: Google, Tesla, and Uber Are
in It to Win It on page 6). Baidu, the owner of China’s largest Internet search engine, says it might
introduce an autonomous car in the near future.2 Apple is in the same discussion with its Titan
project, though its plans remain unclear.
2
Baidu May Introduce Autonomous Car This Year, CEO Says, Bloomberg, 9 March 2015
How Automakers Can Survive the Self-Driving Era
4
Figure 2
Autonomous cars will generate huge economic benefits
$488
billion
Fuel savings
Fuel savings from
avoiding congestion
$158
billion
Total savings from
accident avoidance
Autonomous
cars total
savings:
$1.3 trillion
$11
billion
U.S. market, non-exhaustive
$507
billion
$138
billion
Increased productivity
from autonomous cars
Increased productivity
from congestion avoidance
Source: Predictions for U.S. market, Morgan Stanley research, 2014
Over the next 20 years, autonomous driving will be the culmination of decades of research and
development by many companies around the world, not just the brainstorm of a few disruptive
newcomers. This point comes across early in our study as we interviewed executives from the
traditional OEMs, telecom services and infrastructure providers, and the large network of
suppliers that supports them.
“Premium OEMs will develop their own intelligent cars,” the chief strategy officer at a major
OEM told us. Daimler’s self-driving concept car has four swiveling lounge chairs that allow
face-to-face conversations when the car is in automatic mode. The vehicle is often seen
cruising the streets of San Francisco to give people a glimpse of what vehicles might look
like and how comfortable and seamless connected mobility can be. Another executive spoke
of dedicated academies, run by the OEMs, to help train consumers to become comfortable
with self-driving vehicles.
The companies on the “connected” side of connected mobility are no less ambitious. “We believe
that we will achieve a 10 percent share of the mobility market with drones and commuter
vehicles,” a C-level executive at an Internet media company told us. Another executive remarked
that “drones will be accepted by people who are not able to drive.”
Both the developers of lounge concept cars and the advocates of commuter vehicles and
drones recognize the essence of “connected mobility.” Their solutions mark the first time we
will have seamless integration of connectivity and vehicular mobility, with connectivity being
the force behind the vehicle instead of speed and power.
Some observers believe connected mobility will create a new form of dependency, as people
grow accustomed to satisfying their needs and communication desires from within their
self-driving vehicles. If this is true, then first-mover advantage goes to the first company that
bundles information density, content access, and mobility into a single compelling package.
If one of these companies really does claim 10 to 15 percent of the automotive market, it will
be an important and influential player, not a niche company.
How Automakers Can Survive the Self-Driving Era
5
Google, Tesla, and Uber Are in It to Win It
Google, Tesla, and Uber have had
nothing to do with the traditional
automotive industry but they
are working hard to excite the
public’s imagination about the
prospects of self-driving cars.
Google, for example, has seized
the initiative with plans to take
information gathering from the
prohibitively expensive frontiers
of outer space down to our
neighborhood streets. The
company, determined to transfer
its forerunner and pioneer image
to cars, is testing something
called Google Chauffeur in its
own self-driving car prototypes.3
Google is confident that fully
autonomous driving will be
available to consumers between
2017 and 2020.
Not to be outdone, Tesla claims
its next-generation electric cars
will be 90 percent autonomous,
relying on a control system aptly
named Autopilot. In February
2015, transportation service
provider Uber entered the race,
announcing a partnership with
the Robotics Institute at Carnegie
Mellon University to develop
a self-driving car that Uber can
use in its fleet.4 In theory, this
could enable Uber to make its
own drivers obsolete, not to
mention the traditional taxi
business as a whole.
A recent study describes the
benefits of introducing driverless
taxis.5 Of the many benefits
uncovered, one of the biggest is
price, finding that driverless taxis
can turn a nice profit by charging
$1 per mile, or even less depending
on the size of the fleet. This makes
a driverless taxi very competitive
with public transportation,
especially versus short intra-city
trips or versus park-and-ride
commuter trains.
What outsiders such as Google,
Tesla, and Uber have defined are
the new fault lines between hardware, software, and content in
the automotive industry. Indeed,
drivers will soon take advantage
of augmented reality—using
millions of data points from
multiple sensors to gain a better
perception of their surroundings
(see figure). Augmented reality
can make anything the driver
needs appear more prominently
in his or her visual and acoustic
field, be it an obstacle, a sign,
a path to drive on the road, or even
a warning to slow down or brake.
Rapid technological changes have
given manufacturers and service
providers a unique opportunity
to reshape the mobility and
transportation industry instead
of treating autonomous driving
as a bolt-on, linear extension of
the existing industry.
Figure
Transfer augmented reality pictures within a millisecond
What information is displayed to earn
the autonomous driver’s trust?
Driver
Human-machine
interface
How will the driver interact with the
autonomous driving control loop?
Actuator
Control
unit
Sensor
Environment
Autonomous
driving control
loop
Source: A.T. Kearney analysis
3
Inside Google’s Quest to Popularize Self-Driving Cars, Popular Science, 18 September 2013
4
Uber, Carnegie Mellon Announce Strategic Partnership and Creation of Advanced Technologies Center in Pittsburgh,
Carnegie Mellon University, 2 February 2015
5
5 reasons self-driving taxis are going to be amazing, Vox, 17 March 2015
How Automakers Can Survive the Self-Driving Era
6
The $560 billion question: who will win this race?
It will take up to two decades for fully autonomous driving to emerge. While industry players
have already developed or tested many of the technological building blocks, tough and tricky
legal challenges remain, including new laws on accident liability, on where self-driving cars may
operate, and on who may have a license. Also, new traffic guidelines have to be developed for
autopilot and for fully autonomous driving. The incentives to establish the right legal framework
are high, and executives in our study are confident this framework will develop, probably with
California as the pioneer.
The wake-up call for the automotive industry is that not all incumbent OEMs will get a piece of
the market. OEMs face hard questions as they jockey for position—from thinking about their
value propositions, which core capabilities they need, and which players they should partner
with, to developing business models that offer the best go-to-market strategies and the best
chances to win. These decisions are high-stake and urgent.
We expect numerous OEMs to be major forces behind the growth in autonomous driving. Who
they are and how well they do will depend on their answers to the most crucial and strategic
questions, and how they define the problem and solution of connected mobility. Clearly, given
their long legacies and established infrastructures, it is natural for OEMs to seek solutions that
integrate a connected lifestyle with existing mobility to realize the dream of a mainstream,
self-driving car.
But their headline-grabbing competitors from the Internet world are already defining the
problem and solution from the opposite direction. These well-known game changers with
names like Google, Apple, and Uber want to integrate mobility with an existing connected
lifestyle to realize the dream of a self-driving device. The ultimate device—its size and shape,
its comfort level, what it runs on, the materials it is made from—will emerge from their intense
focus. Designing self-driving devices tailored to lifestyle desires and needs liberates their
thinking. They avoid the trap that autonomous driving must be a series of incremental improvements to cars as we already know them. For example, a Google car at its first stage of autonomous, connected mobility will not travel faster than 25 miles per hour (around 40 kilometers
per hour). At this speed, Google does not need to install the crash safety boxes mandated by
U.S. law. But Google and other newcomers have plenty more to offer. They will secure consumer
dependency with a stream of high-density information in a vehicle that still has a competitive
highway speed. Their focus is meeting the basic needs of the new digital driver who wants to
be always connected.
In contrast, the incumbent OEMs plan to concentrate on systems and integrative functions. So
far, they have been successful with incremental approaches and with bolder ideas such as the
lounge car concept. But can this be the best set of future roles and future thinking for these
OEMs? Or will the demands of the industry be different? How can OEMs—their CEOs, designers,
and product managers—think differently about the lucrative future promised by autonomous
driving and make sure they claim their share of it?
No class of companies has a greater stake in autonomous driving than the car companies.
Yet these same companies are maintaining an equally large stake in the future of the automotive
industry. With a foot in both ponds, is it possible for OEMs to win? There are urgent challenges that
go far beyond the heavy and necessary investments already made in research and development.
How Automakers Can Survive the Self-Driving Era
7
Incumbent OEMs need to think about how to answer five questions and begin making
decisions now:
1. How can OEMs match consumer needs with autonomous driving solutions, while
overcoming skepticism about relinquishing control of the vehicle?
2. How will the market for autonomous driving develop and what will the associated product
roadmaps look like?
3. How will government legislation keep pace with new technologies while also addressing
questions of liability?
4. Which business models will win in the new industry?
5. What role will partner and competitor ecosystems play in autonomous driving?
The challenge for incumbent OEMs is anticipating what that vehicle or mobile device might
look like. Let’s start with the premise that companies such as Apple and Google want to
preserve the seamless always-on lifestyle and recognize that people often want or need to get
from point A to point B. The solution is to take the connected exchange of data and information
that underpins that lifestyle and add mobility. If you think in these terms, what kind of vehicle
or device do you need? Odds are that the end product isn’t a one- or two-ton metal box that
spends more than 95 percent of its time sitting idle.
Markets and Products: Autonomous Driving
Is a Lucrative, Game-Changing Opportunity
The markets
How large and lucrative will the autonomous driving market be? We estimate that the annual
market just for special equipment—onboard control, guidance, and communication systems—will
reach $103 billion by 2030. Mobile apps that facilitate car-to-car telematics and communication
between cars and other entities will account for another $86 billion. Fully autonomous vehicles
with no manual intervention by the driver will start to appear in the late 2020s, and by 2030,
we expect the market for the vehicles alone to reach $95 billion.
The biggest leap over the next decade will be the integration of mobile broadband systems with
robotic drive systems. Companies or alliances of companies that first develop a standard will be
the winners, and their reward will be revenue and double-digit EBIT growth.6
All told, the apps, equipment, and vehicles related to autonomous driving will pull in $282 billion
in revenues by 2030, which represents about 7 percent of the total automotive market (see figure
3 on page 9). The numbers get much bigger from there. We expect the market to almost double
to around $560 billion between 2030 and 2035 and represent 17 percent of the global automotive
market. And this estimate does not even consider ancillary revenues from mobile apps, which
will form the basis of many in-car services. By 2030, traffic management systems will free up
a mind-blowing 1.9 trillion minutes for passengers, most of whom already have a smartphone
within arm’s reach at all times (see figure 4 on page 9). The competition for drivers and
passengers will be fierce.
6
EBIT is earnings before interest and taxes.
How Automakers Can Survive the Self-Driving Era
8
Figure 3
The value of the connected mobility market
Global market for automated and autonomous driving, including related services
($ billion)
17% of total automotive market
558
42
Mobile apps with digital features1
67
Apps and goods with digital and physical features2
Special equipment (for high/full automation)3
7% of total automotive market
Fully autonomous vehicles
189
282
38
48
1
51
83
15
26
13
28
23
29
2020
2025
103
260
95
2030
2035+
Content and software for autonomous driving
2
Telematics features for car-to-car and car-to-x communication and traffic management
3
Accessories for assisted driving, auto pilot, navigation, and more
Sources: IHS Automotive, Berylls Connectivity Compass 2014, Factiva, Just Auto; A.T. Kearney analysis
Figure 4
Self-driving will free up 1.9 trillion minutes of idle time in 2030
Global self-driving minutes*
(billion)
Aggressive surplus scenario
5,102
Moderate scenario
4,410
CAGR = +50
3,740
3,091
2,462
1,852
1,333
898
87
17
70
266
2025
2026
53
213
542
108
434
180
2027
2028
718
1,066
2029
1,482
2030
882
748
618
492
370
267
1,970
2031
1,020
2,473
2032
2,992
2033
3,528
2034
4,082
2035
*Long distance commuting only
Source: A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era
9
While the market will more than double between 2030 and 2035, the growth across categories
will not be uniform. Revenue from special equipment for autonomous driving will peak around
2030, after which technology will start to become a commodity. Revenues from content,
software, and services will exceed revenues from special equipment. Meanwhile, revenues
from the self-driving vehicles will almost triple. So-called derivatives will also play a role in the
market as autonomous driving vehicles are designed for even more narrow purposes and are
well-suited to perform only these given tasks.
The apparent quantum leap in both technology and revenue from 2015 to 2035 will not be a
single leap but a steady and rapid progression through a series of generations (see figure 5).
The transition to fully autonomous will also be gradual and steady. It will follow a technology
roadmap that includes mobile broadband systems, vehicle radar and location systems, and
automation or robotic drive technologies using big data infrastructure management such
as augmented reality. Tapping into this vast potential will require unprecedented levels of
collaboration and cooperation among OEMs, telecommunications firms, media and Internet
companies, and government bodies.
By far the most compelling part of this story is not the forces that give rise to mainstream
autonomous driving. Rather it is the high-stake question of which company, or alliances of
companies, will harness these forces to win the largest and most lucrative share of the market.
It is dangerous to assume that the sheer size of the market will guarantee that everyone is a
winner and that the only difference in success is a matter of degree.
Maturity
Figure 5
Autonomous driving technology will advance in waves
Partly
automated
2015
Highly
automated
2020
Fully
automated
Autonomous
(robotic)
driving
2025
2030
Global
megacity
networks
2035+
Legal
protection
Scoring system
for preventive
driving
Daily legal
protection for
car rental
and car-to-go
Legal protection
based on
mobile devices
Legal protection
for all kinds
of devices
Legal protection
for all kinds
of devices
Technology
First-generation
automation
and control
Second-generation automation,
pooling
Third-generation
fixed-distance
automation
Fourth-generation automated
traffic junctions
Fully automated
traffic flow
management
Infrastructure
Highway
networks
Regional
or national
network
Selected
megacities
Micro-/mobile
city metropolises
Globally
interconnected
megacities
Standards
Internet
standards for
mobility apps
Camera and
image processing
and interfaces
Radio frequency
and interface
standards
Control and
automation
standards
Fully automated
networks/
telematics
Sources: Rinspeed; A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era 10
The products
The always-on consumer requires connectivity. Connected mobility changes the automotive
market in two fundamental ways. First, it creates new service opportunities. The more interwoven the concepts of connected and mobility become the more sophisticated the market
solutions will be—all the way up to connected urban centers with advanced management
systems to optimize the flow of traffic. Second, connected mobility creates new forms of
competition as it opens up the market to connectivity specialists and blurs the lines between
connectivity and mobility.
In this highly automated world, demographic, environmental, and technological forces will
result in many new products and services that will spice up the mobility experience, not just
make it more practical and efficient. We group these products and services into four categories:
mobile apps, special equipment, autonomous vehicles, and infrastructure.
• Mobile apps. The mere existence of mobile apps can alter the role of the automobile. Apps
enable consumers to treat their cars as extensions of their home, school, and office rather
than simply a means to travel between them. Apps can also remotely manage the energy
consumption and security at these locations, including turning on an alarm system, turning
down the heating, or turning lights on or off. Drivers and passengers will use their smartphones or an app integrated into the vehicle.
These consumer-to-car (C2C) apps include entertainment, multimedia, and information.
With autonomous driving, a solo “driver” can watch a live sports event, find a restaurant or a
store, research a leisure destination, or relax and play a game, all without being a distracted
driver. They can stay in touch with friends, family, and colleagues, and it is entirely feasible—
and perhaps very efficient—to conduct a video conference from a moving vehicle while the
vehicle drives itself. Speaking at Nvidia’s annual developer conference in March, Tesla CEO
Elon Musk compared this process to taking an elevator: “They used to have elevator operators,
and then we developed some simple circuitry to have elevators just automatically come to the
floor you’re at. The car is going to be just like that.”7
Apps will integrate the vehicle with a smart home, programming your home from the car or
summoning the car from home. Apps will also provide more flexibility for insurance, safety, and
legal protections. A consumer summoning a pay-as-you-go car can choose an appropriate
package for that day, depending on the weather, the type and length of journey, and the
number of passengers.
• Special equipment. Several OEMs have launched automatic parking assistance, and the
feature is not limited to premium manufacturers such as BMW or Mercedes. Ford and
Chevrolet also have models with hands-free parking. This is only the warm-up act for many
practical features to follow in the form of driver-to-car (D2C) communication.
The next step is automatic driving or “chauffeur mode,” which will offer more comfort for the
driver. This technology already exists in a practical form, as Audi proved in the run-up to the
2015 Consumer Electronics Show in Las Vegas. A team of engineers and journalists traveled
about 550 miles (900 kilometers) from Silicon Valley to Las Vegas in a specially equipped Audi
A7. The drivers engaged the autopilot with a press of two buttons in the steering wheel. The
car then sped up, slowed down, and switched lanes—all without human intervention.8
7
Elon Musk: self-driving cars could lead to ban on human drivers, The Guardian, 18 March 2015
8
Audi Drove This Auto-Pilot Luxury Car From San Francisco to Vegas, Business Insider, 6 January 2015
How Automakers Can Survive the Self-Driving Era 11
Now imagine that same trip on a conventional highway but with many more cars able to
detect obstacles, respond to traffic conditions, and even communicate with each other.
Car-to-car telematics (C2C) improve safety and mileage range, regardless of whether the
car is gas-powered, hybrid, or battery-powered. It adds an important information element
to decision making on the road that humans lack. Even drivers with the greatest skills,
experience, and judgment never know with certainty what the other cars on the road will do.
By communicating with each other to resolve problems on the road, cars travel in a coordinated way and reduce the need for stop-and-go, frequent acceleration and deceleration,
and other abrupt ways that we cope with traffic.
• Autonomous cars. Fully automated driving means drivers and passengers can lean back
for the entire journey as the car finds the best way to get to the destination, including any
required parking or other stops. Self-driving cars can form platoons, or groups of interconnected cars with a fixed distance and speed. They can also swarm, which means the car
gives control to a real-time automated traffic flow management system. That system takes
the demands of a group and its individuals into account to make the best use of available
roadway, whether it is passenger cars or commercial vehicles. We expect partnerships to
spearhead development of several types of autonomous cars, each designed for a narrow
purpose. These include premium high-end lounge vehicles and travel and experience
vehicles, simpler vehicles and taxis for commuting and short family trips, and drones that
can increase information density and capacity and offer options for speed.
• Infrastructure. The powerful communication that enables these previous categories will
require a different infrastructure from that found in today’s highways and cities. Fortunately,
most of this new infrastructure can overlay or work with existing roadbeds and communication
systems rather than replace them. The first step is to ensure that networks, protocols, and
standards exist for real-time interconnection between cars and other entities. The second
step is to install platforms for automated traffic control on highways and in urban areas to
enable robotic driving.
Competitive Landscape:
You Won’t Recognize It 20 Years from Now
To say autonomous driving will reshuffle the auto industry is putting it mildly. When connected
mobility is the umbrella phrase to describe vehicular transportation over the coming decades,
new players—especially from the “connected” side—will want to interact directly with
consumers. OEMs need to selectively claim new touch points with their current and potential
consumers as protection from these threats.
As you might imagine, traditional automotive OEMs are not about to concede the public’s mind
and money to companies such as Google and Tesla. OEMs have no plans to become the shortsighted “buggy whip manufacturers” of the 21st century and have begun to fight back with their
own groundbreaking initiatives. They have altered their branding in anticipation of autonomous
driving and intensified their research and development efforts. Most OEMs have set a timetable
for commercial availability of semi-autonomous driving within the next two to four years.
These initiatives are important as the automotive industry takes the first step along the technology
roadmap that leads to fully autonomous driving. But these initiatives do not necessarily account
How Automakers Can Survive the Self-Driving Era 12
for two inevitable changes in the automotive industry: the value chain, which describes how
players work together to produce a vehicle and take it to market, and the value share, which
describes the sources of the vehicle’s total value and their relative contributions. The following
takes a deeper look at both:
Value chain: from value “pyramid” to “hub and spoke.” The existing value chain in the
automotive industry is best described as a pyramid, with the OEMs at the apex. Under them are
several tiers of suppliers, ranging from raw materials and component suppliers at the base (third
tier) up to systems suppliers and fully integrated production partners (tier 1 and 0.5 tier).
The new logic is neither a chain nor a pyramid, but rather more of a hub-and-spoke arrangement.
The finished vehicle remains at the center, surrounded by indispensable and in some cases
interconnected parts of a wheel: tier-x suppliers, the OEM, IT suppliers, online players, telecom
companies, and device manufacturers (see figure 6). Remove any part, and the wheel doesn’t
roll. The player that controls the customer relationship makes the wheel spin. For decades, the
OEM held that responsibility. That might not be the case 15 years from now.
Figure 6
Autonomous driving will disrupt the automotive industry
Existing value chain
New hub-and-spoke
OEM
OEM
0.5-tier
supplier
Tier 1 supplier
(systems)
Tier 2 supplier
(modules or components)
Tier 3 supplier
(raw materials, intermediate goods, or components)
Tier-x
suppliers
IT
suppliers
Device
manufacturers
Online
players
Telecom
companies
Source: A.T. Kearney analysis
Compounding the challenge for automotive OEMs is that nearly all participants in the other
spokes are multibillion dollar companies with strong research and development teams, regional
or global market leadership positions, and an appetite for large, game-changing growth opportunities. Think Microsoft, Google, SAP, Samsung, Siemens, Deutsche Telekom, and even Twitter
and Facebook. In 2035 to 2040, to the extent we see a pyramid at all, it will be flatter and
broader, with a mix of new and old mega-players, wild cards, and specialists.
Value share: vehicle value undergoes a tectonic shift. As the logic underpinning the
automotive industry changes from a pyramid to a wheel—or perhaps even because of this
How Automakers Can Survive the Self-Driving Era 13
change—the value of a car will also undergo a tectonic shift. This will add to the pressure on
OEMs. Today, the value for an average automobile is 90 percent hardware and 10 percent
software. The future value looks much different. Hardware’s share plummets to 40 percent and
its profit pool shrinks. Most of the value lies in software (40 percent) and content (20 percent),
including the apps that bridge and integrate the hardware and software.
We expect software and content providers to achieve the highest margins. The most likely
candidates to lead this area are technology companies with know-how in apps and operating
systems. Automotive software providers will have less profitability than content providers but
will still have better margins than hardware providers. These software companies could be
technology firms, existing OEMs, or automotive suppliers that recognize the opportunity and
make an early move to capitalize on it. While all relevant technologies exist for autonomous
driving, the ability to integrate car software architecture and speed up software development
must make major leaps forward (see figure 7).
Incumbent OEMs and their suppliers will still dominate automotive hardware because of their
process expertise, but commoditization and a much lower value share will put their profits at
considerable risk. Another crucial point is the growing gap between the life cycles of software
and content platforms, their modules, and vehicle hardware. An individual module may have
a very short life cycle, but modular programming or agile software development could allow
a software platform to have a life cycle of up to four years—twice as long as the life cycle for
hardware, such as semiconductors.
Figure 7
Software for connected cars needs improvement
Integrated car software architecture
Relevant technology
Radar
Radar
Engine
Radar
Steering
Steering
ECU
Brake
ECU
Vision
Suspension
ECU
Central controller
Cabin
Brakes
Safety
Infotainment
Infotainment
Radar
Safety chassis
domain controller
Radar
GPS
Needs
improvement
Cabin
Safety
Steering
Engine
Brakes
Secure and fast software development
Requirements
Gate review
Driveline
ECU
Radar
Daily scrum
• Plan
• Design-buildtest and testdesign-build
High-level
design
Gate review
Detailed
implementation
Gate review
Radar
Verification
Prioritized
customer
requirements
Gate review
Inspired by
large US
military
programs
of the 1960s
Deployment
Product
and release
backlog
Sprint
demo
Sprint
retrospective
Backlog
preparation
Selected
backlog
Sprint
execution
Sprint
backlog
Sprint
planning
Potentially
shippable
software
Customer
validation
Note: ECU is engine control unit.
Sources: Autoliv, Morgan Stanley
How Automakers Can Survive the Self-Driving Era 14
Five categories of competitors
We group the competitors in the market for connected mobility and ultimately for autonomous
driving into five categories: luxury OEMs, middle-class OEMs, low-cost OEMs, tier 1 suppliers,
and wild cards (the disruptive newcomers to the automotive world). Who will survive as this shift
in value occurs? From within the existing automotive industry, we believe the luxury OEMs and
some tier 1 suppliers will emerge as winners, provided they start to specialize in software. The
wild cards may be the biggest winners of all, or at least produce constructive failures that leave
a compelling legacy. In contrast, the middle-class incumbent OEMs are in a precarious position.
In this section, we discuss the prospects and challenges for each of the five categories, along
with our recommended courses of action.
Luxury OEMs: Invest in software, seek alliances
Examples: Audi, Porsche, Mercedes-Benz, Lexus, and BMW
Luxury OEMs face several challenges. They must uphold their technology reputations while also
realizing that no matter how advanced their own technology is, they cannot succeed without
support. In the old pyramid structure, this meant finding and vetting the best suppliers. In the
new hub-and-spoke world, it means finding the right allies and establishing partner networks
and ecosystems. Recent strategic moves indicate companies have begun strengthening their
positions (see figure 8).
Figure 8
Ecosystem partners are strengthening their positions in the value chain
Content
providers and
aggregators
Strategic
moves
Content
is king
Device manufacturers and application providers
threaten access
network providers…
facebook
NAVTEQ
Application
and service
providers
Secret
sauce
Access
network
Not just
bit pipes
ovi NOKIA
QUADRANT
iTunes
OpenWay
Comverge
… forcing them
to expand via
partnerships or
integration along
the value chain
VZ
U-verse
HMV
QUADRANT
Devices
Distribution
Not a
commodity
Last mile
to consumer
BlackBerry
NOKIA
GARMIN
BMW
WebTech
Apple
Itron
Verizon
at&t
Sirius Satellite
Radio
Bell
DirecTV
at&t
BMW
TiVo
WebTech
HMV
Source: A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era 15
Also, luxury OEMs must assess their core capabilities and establish a make-or-buy strategy to
fill any gaps. Daimler, for example, wants to cooperate with Apple on off- and onboard systems
and on applications for easy-to-handle monitors, while BMW wants to collaborate with Apple
on connectivity and human-robotic interface (HMI) applications (see figure 9).
Figure 9
Companies will join forces to create powerful ecosystems
Non-exhaustive
High-tech device
manufacturers
Apple
Audi
Nokia
SAP
Software
providers
Samsung
Toyota
Microsoft
Google
Mercedes-Benz
HRS
Traditional
automotive players
WhatsApp
Online
services
Huawei
BMW
T-Mobile
faurecia
Pandora Audible
Foursquare
Vodafone
Telecom
companies
Vinci
Infrastructure
providers
BOSCH
Volkswagen
Twitter
China
Unicom
Huawei
Renault Nissan
Deezer
Telefonica
Siemens
Groupon
Facebook
Social communities
Source: A.T. Kearney analysis
Of the two, BMW will likely struggle to build strategic alliances on this activity field. If BMW
overcomes some challenges, it could capture first-mover advantage in this competitive race.
The automaker can also create “walled gardens,” proprietary information worlds where its
customers can interact and neutralize the advantage of Silicon Valley firms in this area, or
prevent potentially unfavorable partnerships with them.
Let’s explore the walled garden concept. Imagine OEMs starting their own kinds of “Google
world” where free-spirited millennials find their preferred automotive brand and then subscribe
to a comprehensive mobility plan that lets them experience the whole planet in whatever mode
they choose: on their own, with a robotic driver, or together—all traveling in group mobility
mode. In addition, a premium OEM might offer its own app, an onboard hardware feature
package, or distinctive spinoffs with equipment for autonomous driving (such as a lounge,
adventure, or peer-group transportation concept).
On the other side, non-automotive players continue to change the game—cooperating with
automotive OEMs to achieve the market’s pole position.
How Automakers Can Survive the Self-Driving Era 16
Middle-class OEMs: Deep trouble ahead
Examples: Toyota, Volkswagen, SEAT, and Citroën
Any market has a thriving middle when competitors offer “best of both worlds” combinations
that appeal to a large base of consumers. Such companies build their brand on complementary
phrases such as “affordable luxury.” On the flipside, the middle of the market can be an area that
is neither fish nor fowl and thus has far fewer consumers.
Middle-class OEMs fall into the latter position. Wedged between luxury OEMs and low-cost
OEMs, they will struggle to find a value proposition or key selling point in the era of autonomous
driving. Their target price range is too low to please early adopters, but too high to sustain them
when technology becomes a commodity.
These mid-range companies are also focusing on business-to-business premium offerings,
bundling equipment for connectivity and human-robot interaction. They also want to develop
modules to offer commuters various powertrain options such as e-vehicles and hybrids.
Middle-class OEMs will try to focus on connected services and telematics, filling apps and
equipment hardware packages much like the early adapters (GM, BMW, and Ford) considered
car telematics a core element of their development strategies (see figure 10). To expedite time
to market, mid-class OEMs will want to form strategic alliances with strong non-automotive
partners. And to secure a viable role between the premium and low-cost segments, they will
create a solid and balanced mix of off-board and onboard intelligence, radio frequency input,
and auto-robotic driving.
Figure 10
Early adapters will build car telematics into their strategies
Non-exhaustive
The evolution of connected services
BMW
Firstgeneration
telematics
1996
BMW
Pan EU
telematics
MercedesBenz
Teleaid
PSA
Peugeot
Citroën
Pan EU
telematics
Mercedes-Benz
mbrace
Toyota
safety connect
1998 2000 2002 2004 2006 2007 2008 2009 2010
OnStar
by GM
First
generation
Ford
SYNC by
Microsoft
VOLVO
On Call
BMW
Assist
second
generation
OnStar
and
Toyota
GBook
(China)
HondaAgero
Mazda
roadside
assistance
app
2011
2012
Mini
Connected
VW-Hughes
telematics
(United States)
2013
2015
eCall
mandate
GM
my link
Hyundai
BlueLink
Toyota
Entune
OnStar
FMV
–
Increase in telematics subscribers
+
Sources: Frost & Sullivan; A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era 17
Low-cost OEMs: Invest in software and seek alliances
Examples: Dacia, Kia, Hyundai, and Daewoo
The hub-and-spoke value chain puts a premium on partnerships. No category is in greater need
of such technology partnerships than the low-cost OEMs, which need to compensate for their
limited R&D resources. Their experience with a low-cost production process, and the ingrained
culture that supports it, should enable them to develop low-cost, self-driving cars. If they can
establish positions as the suppliers of automotive drones, they can carve out a niche and be
ready when technology becomes a commodity and infrastructure matures.
Google or Apple might collaborate with low-cost OEMs to offer extraordinary communication
and auto-robotics platforms using tactile Internet and mobile broadcasting systems. For
example, Hyundai BlueLink provides an onboard system plus smartphone plus Web platform
that offers navigation, emergency alerts, protection, communication, maintenance, remote
access, and three packages to choose from with increasing levels of service. If the goal is to find
the best way to pair “connected” with “mobility” to create an appealing and affordable package
for consumers, the leaders might be best served to work with efficient manufacturers of
low-cost transportation instead of making connected mobility a premium offering in all aspects.
Tier 1 suppliers: take over hardware, or specialize in software?
Examples: Bosch, ZF/TRW, Continental, and Schaeffler
Tier 1 suppliers face perhaps the greatest range of options and challenges among existing
automotive industry players. The first challenge is to maintain a competitive position as highquality suppliers. Then they must decide between two strategic directions, both of which are a
fundamental departure from their current strategies. They could move upward in the thinking of
the old pyramid structure and displace OEMs as hardware suppliers. Or they could shift the
focus of their value creation from hardware to software and the components that support it.
We recommend the second option. Tier 1 suppliers should prepare for a shift in value creation
from hardware to software. In the short term, this means making R&D investments in sensors,
devices, and other critical components; in the long term, they need to establish reputations as
software suppliers, not just hardware suppliers.
Wild cards: win or lose, they will be very influential
Examples: Google, Apple, Facebook, Cisco, Uber, and Microsoft
The self-driving car as a consumer product and platform for services opens up the market not
only to current manufacturers, but also to wild card firms. Each wild card brings its own global
strengths to help make autonomous driving part of new mainstream lifestyles. These companies
built their reputations and fortunes on “connected” and relish the opportunity to pair
“connected” with “mobility.”
They have also witnessed, profited immensely from, and in some cases instigated previous
transformations of physical business models into digital ones. Think of music and movies,
established industries where millennials prove an industry can be transformed in the span of a
generation. Movies and music are not something millennials own but rather something they
summon at will to fit their needs, whims, and circumstances. It will be no surprise if they prefer
to summon a car on demand or on subscription, as they do with their music and other entertainment forms, rather than own it. In this spirit, we predict revenues from pay-per-use services
will exceed revenues from optional onboard equipment from 2025 onward.
How Automakers Can Survive the Self-Driving Era 18
Another threat to incumbent OEMs is the potential price advantage a company such as
Google might claim. Driverless cars from Google are not only passenger transportation, but
an ingenious data collection system. If passengers feel comfortable exchanging rich data—
telemetry, pictures, and videos—in exchange for a ride, Google can significantly lower the
barriers to access. There is no reason, however, why the incumbent OEMs cannot either pursue
a similar strategy or neutralize Google’s. This will be a question of their ability to develop core
competencies in data and information or find the right attractive partners to create an
experience for consumers.
Despite all these positive headline-grabbing prospects, the wild cards face challenges. They
need to find a way to expand their existing platforms and networks into consumers' cars and
then anchor them there. Then they need to take advantage of transaction data to earn money
from their value-added offerings.
Most original equipment manufacturers
have set a timetable for commercial
availability of semi-autonomous driving
within the next two to four years.
Another potential outcome is that consumer demand and technological edge will force OEMs
to integrate rival platforms such as Android and Apple’s CarPlay into their cars, essentially
waving the Trojan horse through the gates with full knowledge of what is lurking inside. These
players will not earn their money with the car value itself (the hardware or the engine). They
enter the market with new forms of connectivity, human-robot interactions, auto-robotic
drive, and new telematics. Their attractiveness will derive from the mix of information, media,
mobility, and speed, perhaps coupled with an e-drive concept.
The car will also not be sold over an upfront purchase price. We envision that a Google or Apple
car will be financed by pay per use, fees, flat subscription prices, or licensing. Money will also be
made from short churn cycles on lifestyle apps as well as mobility services and mobile speed
applications, which receive annual, monthly, or even daily updates. These companies calculate
on a profit margin of 10 to 15 percent, but the more highly appreciated features could help them
achieve outstanding, breakthrough margins of around 20 to 22 percent.
Let’s take a look at some of the recent steps of the more prominent wild cards:
Google
• Unveiled a second-generation prototype of its self-driving car project, Google Chauffeur
• Acquired a military robotics maker
• Developed Google Nest for smart-home integration
• Created Android Auto to expand into the car, starting with tactile Internet applications
How Automakers Can Survive the Self-Driving Era 19
Apple
• CarPlay opens the door for Apple’s iOS operating system to be integrated into the car
• The HomeKit interface enables integration with smart home devices
• iCloud hosts a comprehensive database of customer data; expect Apple cars to feature highdensity information and mobility velocity that does not come at the expense of speed
Rinspeed
• Developed the XchangE concept car in cooperation with A.T. Kearney to showcase the
potential of fully autonomous cars (see figure 11)
• Focused on self-awareness to enable advanced features such as cross-car communication
and adaption to a driver’s habits and preferences
MirrorLink
• Formed a cross-industry alliance of hardware and software for handheld devices (LG, Nokia,
Panasonic, Samsung, Sony) and in-car equipment (Alpine, JVC, Pioneer)
• Developed an open interface for connecting users’ smartphones and certified car interiors
Figure 11
Rinspeed’s concept vehicle: a glimpse into the future
Individual connectivity
Group connectivity
Compatibility with docking
and charging hubs
Induction
Examples
People-to-people and
car-to-car communication
Conduction
Car-to-third party connectivity
Real-time availability check,
reservation, and payment
Infrastructure
technology
• Parking lodge
• Car parts: go and load
• Energy net for
loading station
• Rush hour
• Traffic accidents
Board system
Community features
FleetBoard
myBoard
Source: A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era 20
Technology: from Driver-Centric to Connected Mobility
As autonomous driving technology evolves and adoption becomes more widespread, we
envision three lanes of highway traffic—a premium lane, commuter lane, and drone lane. Each
lane corresponds to the product segments with best success prospects, and each is defined by
the nature and sophistication of their embedded technologies.
Imagine standing on an overpass that spans a three-lane expressway. Looking down, you see cars
whizzing by in the left-hand lane. These are the premium vehicles, driving themselves with no
speed limit. What nameplates will these premium cars carry? This is where we have the greatest
certainty over what brands you will see: Audi, Porsche, Mercedes-Benz, and BMW—the same
brands that fill the high-speed lanes on expressways and autobahns today.
The nameplates in the other lanes are far less certain. In the middle lane are the commuter
vehicles, traveling within predefined speed limits. Designed for short, regular trips that allow
passengers to seamlessly continue their always-on lives, these vehicles create many opportunities to trade-off features against price to optimize the vehicles and keep them affordable.
The right lane is the domain of the drones, with speed that varies according to the demands and
destinations at any given moment. The mix here can include all kinds of service vehicles, from taxis
and delivery to specially designed vehicles for shopping, doctors’ visits, and school transportation.
Regardless of the nature of these vehicles, they all fall under two large umbrellas: car-centric
communications and an overarching legal framework. These umbrellas provide the forms,
leeway, and definitions that lay the groundwork for the autonomous driving boom.
The technology umbrella: car-centric communications
The basic technologies to enable autonomous driving either exist today or are being developed
and refined. These are the puzzle pieces that bring autonomous driving to life. But puzzle pieces
are useless unless assembled. This is the challenge that OEMs, wild cards, and their technology
partners face in realizing the full potential of this multitrillion-dollar opportunity.
The future of autonomous driving will have less to do with the mechanics of the vehicle and
almost exclusively to do with forms of interconnection and car-centric communication. This is
the future hardware side of connected mobility. It underscores why alliances and partnerships
are the ideal ways to make autonomous driving desirable and practical. No company has a base
of competencies broad enough to establish a global, 360-degree system for navigation,
control, and automation, built on a stable stationary and mobile broadband network. And no
company will succeed on its own. Groundbreaking partnerships and alliances are mandatory.
The easiest way to appreciate and understand the communication complexity is to look at the
various combinations necessary to make autonomous driving a desirable, practical reality. As
shown in figure 12 on page 22, there are four main communication connection points: consumer
to car (C2C), driver to car (D2C), car to third parties (C2X), and third party to third party (X2X).
1. Consumer to car (C2C)
The first human-car interaction comes when we decide to take a trip of any kind. Today, the four
most common next steps are to get into one’s own vehicle, hail a taxi, use a ride-share service,
or rent a vehicle. In the future, the exchange of data between the consumer and car occurs long
before any of these steps thanks to technology.
How Automakers Can Survive the Self-Driving Era 21
Figure 12
Carmakers are building strategic alliances
Examples
C2C
D2C
C2X
X-smart devices
Onboard vehicle management
Telematics city traffic
• Mixed devices
applications
• Integrated pre-safe
manager
• Multi-antenna
radar technology
• Mobility/automatic/
multimedia app
• Auto-robotic pilot
• Micro-line 360°
camera technique
• Auto-running check
• Intermodal mobility
manager
Apple
Daimler
Google
Intel
Google
IBM
Daimler
X2X
Telematics landscape
• Fast radio-frequency
technologies
• Multi-router
technology
Ericsson
Daimler
Telekom
• HD receiver/transmitter
Google
Bosch
Daimler
Bosch
Apple
Conti
Robotics and
automation landscape
• External
computer system
Google
Tesla
IBM
Daimler
• Big data speed
• High-speed car xCPU
Bosch
Toyota
Siemens
Google
•
Intel
ARM
ABB
Notes: C2C is car to car. D2C is driver to car. C2X is car to third parties. X2X is third party to third party.
Source: A.T. Kearney analysis
Customers will have access to their personal data either via cloud services or physical and
personal storage devices via OEM-promoted Internet access suites. Storing, providing,
managing, and analyzing that data helps cement the relationship between the customer and
the owner of the gateway. This helps explain why ride service giant Uber wants to convert its
fleet to self-driving cars. In February, Uber confirmed it has entered a strategic partnership with
the Carnegie Mellon University Robotics Institute to develop self-driving cars.
When a rider orders a rental car and the car arrives, the car needs to know it has reached the
right person. The rider will also need to synchronize his personal data with the car, whether it
is his own vehicle or a rental. This includes everything from how the rider obtains the vehicle,
interacts with the vehicle in terms of data exchange (payments, destinations, and other information), and returns the vehicle when the trip is completed.
Today, Daimler has 10 applications that can be preinstalled on a black box or via the Internet.
Figure 13 on page 23 illustrates its FleetBoard cockpit showing both the operator and driver
view. These cover aspects such as car check control, preventive maintenance, energy usage,
and automatic control of select services to avoid the frustrating, unexpected downtime of a
vehicle breakdown. Customers become sensitive to the same issues that manufacturers know
intimately, namely that unplanned downtime is their costliest hazard.
How Automakers Can Survive the Self-Driving Era 22
Figure 13
Daimler’s FleetBoard Cockpit oversees transport, time, and maintenance tasks
FleetBoard Cockpit
Operator view
Fleet management system
FleetBoard
Fleet
usage
Vehicle
status
Driving
assistance
License purchase:
$2,300
Lease plus purchase:
Global apps
Fleet
routing
Remote
diagnostics
Service
function
$1,100 + $115 per month
Pay per use:
19 cents/min
Driver view
Car management system
Vehicle monitor
myBoard
Diagnostics
Driving
assistance
Vehicle status
Service
License purchase:
$1,700
Lease plus purchase:
$570 + $58 per month
Pay per use:
6 cents/min
Source: A.T. Kearney analysis
The solutions include hardware-software connectivity with the vehicle’s main control board.
These assume standards for tactile and stationary Internet and proper interfaces such as
internal and external docking stations for smart devices. We can imagine that one day keyless
access and keyless operation will be the norm as the whole concept of physical driving and
vehicle management is handled by a smartphone.
Our hypothesis: OEMs will propel development of smart devices, set future smart technology
standards, and offer in-house hardware and software modules.
OEMs will figure out how to integrate and apply Internet access and power as well as how to
organize and manage data storage. The ability to accumulate historical information—about
vehicle performance, its tendencies under certain conditions, demand fluctuations on certain
highways and city streets, and demand for additional services such as parking and ridesharing—represents a tremendous advantage for the company that owns the data and can
analyze it to achieve greater efficiencies, optimize services, or offer riders more reliable, datadriven, high-quality services. These include Daimler’s offer of the means to connect car
telematics with traffic telematics.
Our hypothesis: Each OEM will collaborate with telecom companies or other partners to
develop its own proprietary Internet brand and applications with a focus on all-encompassing
consumer contact and support. Cars will be equipped with powerful onboard computers for
mass data processing and storing.
How Automakers Can Survive the Self-Driving Era 23
The flipside of this power is privacy. Autonomous driving with cars on demand will be one of
the most information-intensive activities consumers have ever undertaken. OEMs will have
unrestricted access to the data generated in the context of autonomous driving—starting with
personal basics such as address, bank account, and billing information as well as the telemetry
data from a trip, including route and destination. Cars collect data much the same way the black
box on an airplane tracks and records hundreds of parameters. The amounts of data transmitted
in real time will be massive. Potential cyberattacks will lead to additional measures from the
OEMs to safeguard their flow of data. Autonomous driving will be an irresistible target for hackers.
2. Driver to car (D2C)
Human-car interaction also begs the question of control. What will be the center of intelligence
when autonomous driving becomes commonplace? For more than a century, the sole source
of intelligence on the highway has been the driver. Think of all the stressful decisions a driver
has to make on a trip into a congested city: Is it better to take the highway and pay a toll, or
take a back road? Do you switch routes if you hit unexpected traffic? Is it better to park at the
big garage or the small lot?
The ultimate question is how to handle the human-car interaction most efficiently and what information from the outside world is crucial. OEMs will need their own radar and interface capabilities
provided by satellite and drone service providers and also computing centers, as discussed on
page 25 in the section on C2X communication. We expect control of the car to be denied to a
driver in statistically error-prone situations, as discussed in the legal umbrella section on page 28.
Every car will carry one central data system with interfaces for OEMs, car sensors, other cars,
the driver, local hubs, satellites, and mobile networks. Figure 14 provides an example of a D2C
situation in which the consumer starts an autonomous driving journey with a rental car that is
under 360° multidirectional control and surveillance. Some of the state-of-the-art sensors—
including autonomous parking, fuel-efficient driving, and crash avoidance—are already
available from a range of OEMs and not just in the premium segment.
Figure 14
A consumer’s autonomous driving journey with a rental car
Example
Driver-to-car system
Long- and
short-rang
traffic
Vehicle-driver
intracommunication
board
Broadcasting
services
Personal data:
cloud or portable
storage
Daily insurance
menu bar
Navigation,
Payment
documentation, at commerce
and recording
mobility
Source: A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era 24
Sensors and transmission technology come into play whenever cars encounter objects, such as
other cars, bicyclists, pedestrians, and natural or man-made barriers. They are essential for
crash avoidance and autonomous parking, and they must perform to extremely high standards.
Rational or not, society will hold cars driven by computers to a much stricter safety standard
than cars driven by humans. In fact, the goal for autonomous driving is zero failure.
The question now is which OEMs and company alliances will drive these developments forward.
It will be a massive undertaking. Manufacturers need to improve the performance and placement of sensors to ensure that cars receive and transmit precise information. They need to
address basic questions of communication between cars. When cars drive themselves, it is the
car itself that needs sufficient real-time communication to understand what other drivers are
doing. The greater the ability of the car to sense its surroundings, recognize other cars in front
or in back, process that information, and exchange signals with neighboring vehicles, the more
reliably it can make decisions such as switching lanes, slowing down, or taking an alternative
route. Cars will have far more information for decision making than humans ever could.
Also, the practical and technological aspects of the handoff between human and machine must
be defined. The most common instance will probably be in some emergency situations when
humans need to work with the vehicle or assume control themselves. One dimension here is the
direct transmission of information. Drivers will need to connect with the car either directly via a
mechanical interface or indirectly by using their smart devices for seamless lifestyle integration.
However, drivers will always have an option to take manual control of the vehicle if a critical
chain of events occurs in certain traffic situations.
Our hypothesis: OEMs will enter partnerships with telecommunications and hardware and
software suppliers to co-develop next generation technologies. They may also independently
push for next-generation technologies, such as the development of high-performance
nanochips to ensure a trouble-free autonomous driving experience. OEMs will also offer
combined hardware and software solutions around smart device technology and data. Finally,
OEMs in developed regions will rent capacity for computing, monitoring, maintenance, and
surveillance. In other regions, they will build or facilitate building of the infrastructure.
As an example, imagine that BMW takes over the information flow in the Munich metropolitan
area. We might see OEMs running the telematics centers in cities or urban areas and offering
their own fleet of vehicles, as well as renting access to the whole infrastructure and data
systems. This is another situation where a strategic partnership is essential, as the participation
of a Cisco or an IBM in such a venture is extremely advantageous.
The sophisticated support systems built into today’s cars provide some intelligence and help
with tedious decision making. But all of the stakeholders in autonomous driving, especially
consumers, face a fundamental question: what will the smartest part of the future system be?
Will the car of the future be smarter than its passengers? Or even more futuristic and ominous:
will the entire traffic management system be smarter than both?
3. Car to third parties (C2X)
Communication and physical infrastructure are necessary for cities and their residents to make
the most of autonomous driving. From the standpoint of commuters and residents of megacities, the challenge is to get cities connected to cars and to centralize information gathering and
traffic management. For vacationers, business travelers, and logistics companies, the challenge
is compatibility, or what is called “roaming” for mobile telecommunications. Cities such as
How Automakers Can Survive the Self-Driving Era 25
Berlin, Shanghai, Singapore, and Austin are changing to more intermodal mobility-based
infrastructures in which they are becoming smart cities with the following characteristics:
• Sustainability measures and dashboard
• Management of tenants and city services
• Control of multiple providers and city operations
• Advanced monitoring systems for sustainability
• Complex transport and logistics systems
• Free zone administration
The idea is for cities to develop their own micro-mobility centers so vehicles can move seamlessly
from city to city while staying connected (see figure 15). These could be vehicles with consumers
who made their own ad hoc decisions to travel, or they could be autonomously driven electric
vehicles for passengers or goods moving between cities on predefined routes. Someone needs
to ensure compatibility, which means not only the integrity of the communications to support
the vehicle on the road, but also the personal and financial side. Cities and regions will want to
collect taxes or tolls, and service providers—telecom, insurance, entertainment, Internet media,
software—will want to track usage and charge for their services, all in real time via subscriptions,
flat rates, or a payment bonus accrued through usage, miles, and other loyalty programs that
can prove to be a powerful incentive for consumers.
Cities can capture a significant competitive advantage from these services, including attracting
anyone who wants to enjoy the virtues of city life. OEMs have a strong incentive to compete
in this area as well. In fact, they have an opportunity to enter this area in much the same way
Figure 15
Micro-mobility centers keep vehicles connected from city to city
Example
Los Angeles
Microcity
Park
and ride
“Swarming”
tour buses
Satellite towns
Finanacial
Traffic
district
management
Shopping
Logistics
system
Air travel
New York
Source: A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era 26
outsiders such as Google, Apple, and Uber want to enter a capital-intensive industry (vehicle
production) in which they have no experience. The company that provides a customer with
a vehicle needs to ensure seamless, fast, and reliable communication along with compatibility
and autonomous driving within and across two similar telematics landscapes. Toyota, one of the
leaders in this challenge, is running a test model in Nagoya, Japan.
Our hypothesis: OEMs will invest in regional telematics infrastructures or pay a fee to ensure
full compatibility of systems. They will compete for leadership by developing proprietary mobile
broadband frequency with global coverage—either in partnerships or on their own—and will
build collaborative partnerships to run autonomous driving across cities and regions.
Another big wave of the future is the tactile Internet and mobile broadband, two technologies
that can accelerate the transition to self-driving cars. Instead of asking governments and
taxpayers to retrofit thousands of miles of existing highway with cables and sensors, OEMs can
pack the necessary sensing, guidance, control, and communication equipment into the cars.
The emergence of D2C and C2X communication—facilitated by the tactile or mobile Internet—
means the roads can remain “dumb” while the vehicles collect and exchange information to
guide their passengers through traffic.
Connected mobility within and between cities is important. Consider how navigation, control,
and automation work on a global scale with all the competing players, systems, and infrastructures. Connected cars will travel in three categories: premium lane (high intelligence),
commuter lane (shared intelligence), and drone lane (external intelligence). Everyone will want
a piece of this lucrative business and will think their contribution is the one that will govern
the customer relationship.
4. Third party to third party (X2X)
We conclude this section with X2X, or communications between third parties that are relevant
for cars on the road. Let’s say a customer—perhaps an individual or a logistics company—wants
to make a trip from Portugal to Russia via autonomous driving. Such a trip would include
changes in language, speed limits, traffic laws, and driving habits, along with changes in
measurement systems if the trip is from the United Kingdom (miles) to France (kilometers).
Autonomous driving between different telematics and Internet landscapes adds a thick layer
of complexity. This will require a new level of collaborative effort among countries and industry
to create an infrastructure that enables 360-degree global steering, robotics, and automation,
relying on external and internal intelligence. This infrastructure will need centralized and
decentralized hotspots with local IT, data, and mobile broadcasting systems interconnected
with stationary networks or hotspots. From the perspective of autonomous driving, country
borders could be replaced by telematics landscape borders.
Various players in the autonomous driving market will figure out how to combine internal (car)
and external (outside world) intelligence globally with different standards, IT systems, and
digital infrastructures. Figure 16 on page 28 highlights some of the trends and partnerships that
are making X2X communications a reality.
Our hypothesis: OEMs will develop proprietary mobile broadband frequencies with global coverage and that interface to their existing infrastructures. For a fee, cars will dock into the local traffic
control and information system. OEMs will claim the premium segment for autonomous driving
across regions, while the mass market remains up for grabs among many diverse players: software
and hardware manufacturers, service providers, private equity firms, and even cities and regions.
How Automakers Can Survive the Self-Driving Era 27
Figure 16
Technology grid partnerships are creating value
High
Low
Trends
Content
providers and
aggregators
Application
and service
providers
Content is king
Secret sauce
Monetizing
increasing
demand
for content
has been
challenging
Transforming
data and
content into
applications
creates powerful players
NAVTEQ
CustomWeather
Access
network
Devices
Distribution
Not just
bit pipes
Not a
commodity
Last mile
to consumer
Control over
device makers
and content
providers
is diminishing
Investing in
branding and
partnering
to avoid
commoditization
Will static players
have a bigger
role or remain
sales and service
channels?
NAVIGON
at&t
MAGELLAN
Best Buy
telenav
Sprint
GARMIN
Target
T Mobile
TOMTOM
Walmart
Value created today
Auto
Smart
navigation
trafficmaster
Fry’s
DirectTV
Auto
telematics
DRIVECAM
QUADRANT
OnStar
ATX
Personal
Smartphone
applications
CNN
Google
Y!
iTunes
The Weather
Channel
Verizon Wireless
HUGHES
dish NETWORK
DRIVECAM
GM
GreenRoad
BMW
Best Buy
Sirius Satellite
Radio
tyco Fire
& Security
TXU Energy
Toyota
NOKIA
at&t
iPhone
Sprint
Verizon
conEdison
T Mobile
Best Buy
Home
Home
networking
HBO
Comcast
XBOX
Verizon
Warner Brothers
XBOX LIVE
Motorola
at&t
Sony BMG
Comverge
Sensus
DirecTV
Comcast
Assisted
living
None
ADT
icare
AMAC
tyco Fire
& Security
ADT
Source: A.T. Kearney analysis
The legal umbrella: an overarching framework
The legal challenge facing autonomous driving lies in changing and supplementing existing
laws and regulations that never foresaw the advent of self-driving vehicles. No matter how
strong the desire to speed up the transition, it will happen only as fast as several important and
powerful pieces are put together over the next two decades, including technology, standards,
infrastructure, legislation, and consumer readiness. Further, no one company can make this
future a reality on its own.
While the first movers in this market will likely reap the greatest rewards, the market will require
support from political systems to update laws and regulations, make transmission bandwidth
available, and potentially approve infrastructure funding so driverless cars can take to the
highways in large numbers. The following are the main legal issues to consider:
How Automakers Can Survive the Self-Driving Era 28
Licensing. Autonomous driving will extend the age as younger people, the elderly, and the
disabled can ride as passengers. This is one of Google’s mobility approaches. Governments will
need to create licensing and permit systems. Depending on the technological maturity and
cultural acceptance, autonomous driving may demand more or less skill from the driver.
Implications: During the adoption phase, expect to see incremental adaptations in driving
schools and new types of permits, and prerequisites for more skills from drivers until they grow
accustomed to the experience.
Data ownership. Autonomous vehicles generate vast amounts of metadata. The question is
who owns that data. With many parties having a stake in such data, conflicts of interest will be
unavoidable, putting consumers’ privacy at risk. The list of potential stakeholders is long. OEMs,
insurance companies, financial institutions, Internet firms, and law enforcement agencies, to
name a few, will all try to gain access to the metadata generated by autonomous vehicles and
the surrounding infrastructure. For example, OEMs can leverage their “data provision” position
to help insurance companies increase their profitability. Costs are reduced as data on driving
behavior is used to assess a consumer’s risk profile, and revenues can be generated via data on
customer segmentation, value-added services, and customer loyalty (see figure 17).
Figure 17
Four ways insurers can use data to improve profitability
Reducing costs
Generating revenues
Driving behaviors
Improve the risk profile
Client
segmentation
Innovative valueadded services
Offer incentives for the
autonomous driving portion
of a customer’s driving
behavior (based on fewer
human driving errors)
Expand into new areas;
be a first mover in
autonomous driving
by offering a unique
selling proposition
Add new services
close to the core
business of insurance
Customer loyalty
Reduce customer
turnover with additional
services and tailored
premium offers
Source: A.T. Kearney analysis
Yet data ownership has a murky and perhaps sinister side, as well as applications that range
from benign to beneficial. For example, smart device designers and manufacturers such as
Apple, Samsung, and Google use code programming to push access to end users. But this
restricts the major carriers—Vodafone, Deutsche Telekom, Sprint—from access to their existing
license customers. Further, high-tech and broadcast technology companies that want to add
more end consumers are reluctant to partner with the venture companies or other financial
partners whose cooperation is necessary to buy an Internet media company. Time is running
out even for the strongest industry giants.
The pooling of individual data for analysis purposes can make a smart system even smarter, as
traffic control systems recognize tendencies, make evidence-based risk assessments, and
optimize everything from time to energy usage in a collectively powerful way that individual
drivers never could on their own.
How Automakers Can Survive the Self-Driving Era 29
Implications: OEMs will try to build “walled gardens” around their respective ecosystems and
monetize them with proprietary interfaces.
Liability. If a crash involving one or more autonomous vehicles occurs, who is liable? Legal
questions about responsibility (driver versus vehicle) will inevitably arise, involving executive as
well as legislative powers. While the list of companies that want to participate in data ownership
is diverse and long, no one wants to be on the liability list. But someone has to pay.
Implications: The first requirement is that governments set boundaries to the driver’s scope of
responsibility. In the event of a crash between two autonomous cars and no obvious guilt on
either side, we anticipate that insurance companies will have to split the damage. OEMs will
make every effort to be free from any liability, even when a situation rules out human error.
Life or death. If an autonomous vehicle enters a situation where a crash is perceived as
unavoidable, the vehicle might have to make a life-or-death decision in a split second. This
raises an important ethical question: can a machine make such a decision? Even more extreme
is a situation in which either a traffic control system needs to determine the least lethal crash
scenario or the cars need to resolve it among themselves.
Implications: Algorithms will likely be developed for choosing the lesser evil—that is, choosing
the option with the fewest or least severe casualties (see figure 18). OEMs will need to seek
independent certification by third parties to ensure accurate functionality and avoid legal
disputes, such as the Toyota anti-lock brake software recall in 2009-2011. It may also be possible
for consumers to pay to reduce their crash probability. We expect some form of override so the
driver can disrupt the ride in order to avoid or reduce the impact of a traffic accident.
Figure 18
Autonomous driving requires sophisticated communications
Technologies flow in parallel
Positioning or
navigation:
encoded
information
packages
Short
distance
Self-steering traffic junctions
10 km
100 km
100 m
GPS1
2
?
The cockpit:
information
volume for
steering and
automation
in the Internet
Sensor
technology
Cable
3
Big data volume
1
1
Camera and
imaging
technology
Mobile
router
Car and
human
Stationary
router
?
Stopped car
Rental car
Frequency technology such as LTE, GSM, and multi-user
Source: A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era 30
Answering the Tough Questions:
OEMs’ Survival Depends on It
Incumbent OEMs will need to answer some vital questions with a combination of care, urgency,
and decisiveness. Decisions must be made about how to harness the forces discussed in this
paper, from the forces behind autonomous driving to the markets and products that will define
the industry and to understand the competitive landscape, especially the threats and opportunities. Let’s go through the seven questions discussed at the beginning of this paper:
1. How can OEMs match consumer needs with autonomous driving solutions, while
overcoming skepticism about relinquishing control of the vehicle?
In a world of autonomous driving where connected mobility is the core consumer benefit, OEMs
will need well-crafted messages and the right value propositions in the areas that appeal to
always-on consumers who want seamless access to information, communications, and entertainment. However, this seamless access is only one virtue. We expect other products to evolve
to support driving that is autonomous, efficient, stress-free, and dynamic (see figure 19).
Autonomous driving creates environmental and cost benefits, including less toxic emissions
and better fuel efficiency. In addition, there are tremendous savings in travel time thanks to
features such as automated traffic flow management and automated parking. Safety and
security improve as these vehicles reduce human errors from factors such as inattentiveness
or falling asleep.
Figure 19
Products will evolve along four dimensions
Third party to third party
Car microcity
Intercity telematics
Platoon
telematics
Highway
drive
Off-road
chauffeur
Off-road
assistance
assistance
Off-road
Parking
assistance
garage
Platoon assistance
chauffeur
2026
Premium
Stress-free
Auto
robotic
drive
Highway
2022 copilot
X-ing
assistant
Parking garage
autopilot
Valet—
parking
via smart
phone
Automatic
chauffeur
Premium
cushion drive
2018
Automatic all-wheel Swarm
Off-road
Cross-line
Parking- Road Driving school
routing
brake system
assistant
assistance
parking
assistance lane
Wheel to wheel
eCall
Remote 3D
assistant
top view
Track line
360° Dynamic
Racing trainer
assistance
up- and
passenger car
360° vehicle U-turn brake
downhill
Automatic
pre-safe
Parking lot
system
Night
drive
4-wheel
assistant
visions
drive
Top view
Safety
180°
Active radar
Auto
FLA-night
360°
cocoon
identification
speed Dynamic pilot 1
vision 360°
360° automatic
slow
Front side
distance control down
view 2
Proactive
Auto dynamic
Car
PSC
Remote
All-side
pilot
2
telematics
control
collision warning
24/7/360°
Safety
Efficient
Driver to car
Consumer to car
City
Autonomous
Premium
Dynamic self-drive
Car to third party
Sources: Rinspeed; A.T. Kearney analysis
How Automakers Can Survive the Self-Driving Era 31
In the long run, reliable algorithms and mature technology will prove to be superior to human
drivers. However, incumbent OEMs will need to help make that case by demonstrating that
autonomous driving can transform the industry—and then take credit for it. No OEM should
concede the headlines to the wild cards.
Consumer skepticism can be overcome by OEMs that encourage early adopters to experience
autonomous driving. This will create a hype effect and show that the benefits do in fact outweigh
the fears. In fact, OEMs can draw lessons from the early days of air transportation and how most
people have now grown accustomed to the safety, reliability, and benefits of air travel.
In addition to the enormous societal benefits of autonomous driving, individual users will also
improve their lives by giving up control of their vehicles. As discussed on page 4, autonomous
driving will significantly improve safety and give consumers more time for other activities.
2. How will the market for autonomous driving develop, and what will the associated
product roadmaps look like?
Until 2035, the market will grow up to $550 billion. In our scenarios, the largest segment is
fully automated vehicles. However, retrofit and specialized equipment such as auto pilot and
enhanced navigation for semi-automated driving will also be significant. The market for
autonomous driving-related apps and other digital features will also grow but not as much
as the vehicle market itself.
Autonomous driving is impossible without integrated communication. Much of the technology
already exists or is imminently feasible. However, having the pieces to the puzzle does not
mean the picture is clear. The company—or the alliance—that puts together the most compelling picture will have the advantage. In particular, we are talking about the three types of
communication technologies for transmitting massive amounts of data in real time:
• Short range: sensors with indisputable reliability (ultrasonic, infrared, radar, and real-time
image processing)
• Medium range: cellular networks with gapless coverage, bandwidth, and latency (UMTS/3G,
LTE/4G, and eventually 5G and 6G)
• Long range: satellites with guaranteed availability and coverage (GPS/GLONASS)
The emergence of services and the connected, always-on consumer are already initiating new
business models, and this will escalate in the years ahead (see figure 20 on page 33). Large
new revenue sources will come from two sources:
• Giving customers in autonomous vehicles value-adding services such as city guides, local
deals, automated parking, and traffic jams in exchange for their personal data
• Monetizing customer data using a network of third-party content and service providers; by
2025, we expect pay-per-use service revenues to outperform optional equipment revenues
3. How will government legislation keep pace with new technologies while also addressing
questions of liability?
The market will require support from political systems to update laws and regulations,
make transmission bandwidth available, and potentially approve infrastructure funding so
driverless cars can take to the highways in large numbers. If a crash involving one or more
How Automakers Can Survive the Self-Driving Era 32
Figure 20
The connected ”always on” consumer is driving new business models
Work
Mobility
Commerce
Social
High
Degree of
connectedness
Individual
Home
2013
Low
Customer groups
2020
Remote vehicle diagnostics
Adaptive cruise control
Automated crash notification
Parking assistance
Emergency assistance
Automatic tolls
Smart navigation
High
Degree of
connectedness
Intercity
logistics
Low
Commercial
Storage
2030
Driving performance monitoring
Vehicle infrastructure integration
Fuel-saving navigation guidance
Pay as you drive
Blind spot and lane departure warning
Concierge services
Collision warning system
Night vision
Nearest fuel and fuel prices
Real-time traffic
Security tracking
Telematics
City
delivery
Illustrative
Driving assistance
Secure automated physical access
Intermodality
Automated transport
Smart routing
Ad hoc deliveries
Shared last-mile infrastructure
Smart closed-circuit TV
Vehicle tracking
Smart traffic lights
Traffic video control
Flexible
traffic lights
F
Smart tolling
Commercial lanes
Traffic systems
Infrastructure systems
City security
Source: A.T. Kearney analysis
autonomous vehicles occurs, who is liable? Legal questions about responsibility (driver
versus vehicle) will inevitably arise, involving executive as well as legislative powers. While the
list of companies that want to participate in data ownership is diverse and long, no one wants
to be on the liability list. But someone has to pay.
The first requirement is that governments set boundaries to the driver’s scope of responsibility. In the event of a crash between two autonomous cars and no obvious guilt on either
side, we anticipate that insurance companies will have to split the damage. OEMs will make
every effort to be free from any liability, even when a situation rules out human error.
Autonomous vehicles generate vast amounts of metadata. The question is who owns that
data. With many parties having a stake in such data, conflicts of interest will be unavoidable,
putting customers’ privacy at risk. The list of potential stakeholders is long. OEMs, insurance
companies, financial institutions, Internet firms, and law enforcement agencies, to name a
few, will all try to gain access to the metadata generated by autonomous vehicles and the
surrounding infrastructure.
How Automakers Can Survive the Self-Driving Era 33
4. What business models will win in the new industry?
In the “competitive landscape” section, we examined each segment in the industry under its
traditional business model and predicted the challenges and futures that lie ahead. We
concluded that the premium segment will maintain its technological lead by cooperating with
hardware manufacturers. Companies in the low-cost segment can succeed by developing
inexpensive autonomous drones or making their efficient production platforms available to the
wild cards. Companies in the middle-class segment face the greatest danger, for as their
products lose their emotional appeal a collapse in demand is a very real possibility.
In addition to the always on consumer, incumbent OEMs will need to consider other aspects
when determining their future business models:
• Consumer control. In mobile communications, consumer access is owned by device manufacturers with strong brands and access network providers trying to keep their stakes. This is in
sharp contrast to today’s business model in which OEMs control the customer relationship.
• Branding. Most important for device and content providers is the ability to differentiate an
otherwise commoditized offering. As application and service providers create strong brands
through “killer apps,” how can the incumbent automotive OEMs compete?
• Data ownership. Customer transaction data is traditionally controlled by access network
operators; application and service providers can capture some data on user profiles, but its
value is to advertisers. How much data will OEMs be able to capture, control, analyze, and
use to their advantage?
• Privacy and security. Personal data theft and privacy invasion through location data are
significant concerns. Yet the divulgence of personal data could become a primary means
for companies such as Google to lower the price to use or “own” their vehicles. That same
willingness to reveal data underpins many Internet and telecom business models. Can the
incumbent OEMs play this same game?
• Intellectual property. IP helps define industry standards and protocols that establish a base for
interoperable systems and devices. Intellectual property issues become even more important
as players try to gain control of the value chain. For example, insurers with usage-based insurance (UBI) programs say that access to more data will allow them to segment and pool drivers
more effectively than using traditional risk-pricing models in which customers are categorized
into a few segments based on static parameters such as sex, age, and location. With UBI
customers are categorized into detailed segments based on new parameters, including speed,
number of hours driving, miles, and location (see figure 21 on page 35). Armed with more
detailed data, insurers are able to open up new areas of business and propose custom tariffs
to customers.
5. What role will partner and competitor ecosystems play in autonomous driving?
Partnerships are essential to success in autonomous driving. No company has the capabilities
on its own to cover all of the key elements of a connected mobility experience. Given the diverse
capability requirements across the value chain (R&D for application providers, capex for access
providers), partnerships are vital to providing an end-to-end connected solution.
The first OEM to build a compelling partner network will secure the pole position and have the
best chance to lead and win the race. The traditional value creation pyramid will change into
How Automakers Can Survive the Self-Driving Era 34
Figure 21
Opportunities are opening up for players outside of the automotive industry
Example
Usage-based insurance
Traditional risk-pricing model
Potential new pricing model
• Age
• Declared distance
• Number of driving hours
• Maintenance
• Sex
• Garage
• Time
• Parking
• Driving years
• Claims
• Distance and location
• Weather conditions
• Declared usage
• Velocity or limit control
Categorizing customers in a few segments based
on static parameters such as sex, age, and location
due to the limited availability of data
Categorizing customers in detailed segments
based on new parameters such as speed
and sprinting
Source: A.T. Kearney analysis
a hub-and-spoke network structure, which creates opportunities not only for OEMs but also for
their suppliers and large, established companies from outside the industry.
To maintain their customer relationships, OEMs must change how they interact with customers,
moving from transactional interactions to a process of permanent digital interactions. Because
the customer is “always on” an OEM must likewise be always connected with the customer.
Connectedness is one reason why OEMs need to expand their scope beyond cars to other
areas, including infrastructure.
Who’s Who in a Race to the Future?
No one knows for sure which players will dominate in the race for autonomous driving and
which ones will falter. But we do know that the formation of strategic alliances and the choice of
core competencies are the most urgent decisions incumbent OEMs face. The companies
seeking a stake in this market—whether inside or outside the industry—are well-known global
leaders with large customer bases and in many cases very large war chests. The more alliances
that begin to form, the greater the risk that some companies will be left behind.
Several plausible scenarios reflect both the stakes involved and the kinds of partnerships that
will form to make autonomous driving a self-fulfilling prophesy. Imagine the following
partnership scenarios:
OEM and telecom. Daimler partners with Nokia to gain access to connected end customers.
Daimler offers its proprietary world of data and information via a smartwatch under its own
brand name.
OEM, telecom, and Internet. BMW and Vodafone cooperate to make the first BMW Microcity
Munich a reality. The exclusive and proprietary network includes a BMW tablet, manufactured
in cooperation with Amazon.
How Automakers Can Survive the Self-Driving Era 35
Infrastructure and Internet. Cisco wins the public bid to build the 5G and 6G network in
California by 2017 or 2018. It begins a close cooperation with Google to create Google’s own
network and infrastructure frequencies.
Internet and wild card OEM. Google buys or partners with Tesla and gains access not only to
electric vehicles but also to operating models for battery-powered homes.
OEM and infrastructure. GM cooperates with IBM or General Electric to provide complete
infrastructure and connected mobility solutions for U.S. megacities.
OEM and services. Chinese auto manufacturer Chery launches a car-sharing service in
Shanghai and Hong Kong.
These scenarios are intended to fuel a discussion about who can partner with whom and how
that creates more value for consumers than other forms of partnerships. Forward-thinking
OEMs will act now. The clock is ticking.
Authors
Michael Römer, partner, Munich
[email protected]
Steffen Gaenzle, principal, Stuttgart
[email protected]
Christian Weiss, consultant, Munich
[email protected]
The authors wish to thank Sebastian Schoemann, Thomas Hausmann, Arndt Heinrich, and Christian Sames
for their valuable contributions to this report.
Postscript: Implications of Fatal Collision of Tesla Model S
In early May 2016, a 40-year-old man was killed in northern Florida when the Tesla Model S he was
driving collided with an articulated truck that turned left in front of his path. The car was running
Autopilot, a beta program for semi-autonomous driving. The exact causes of the collision are
still unknown, as investigators from the US National Highway Traffic Safety Administration, the
US Senate Commerce Committee, and Tesla continue to gather and analyze data.
The tragic accident in Florida highlights that the technology for fully autonomous vehicles, in
which “drivers” can completely disengage from events on the roadway, is still several years away.
Furthermore, it will reignite the debate on the need for clearer legislation, as well as discussions
around legal liability.
We believe, however, that this misfortune will not fundamentally alter the timeline or the business
models we envision for the self-driving era.
August 2016
How Automakers Can Survive the Self-Driving Era 36
A.T. Kearney is a leading global management consulting firm with offices in more
than 40 countries. Since 1926, we have been trusted advisors to the world's foremost
organizations. A.T. Kearney is a partner-owned firm, committed to helping clients
achieve immediate impact and growing advantage on their most mission-critical
issues. For more information, visit www.atkearney.com.
Americas
Atlanta
Bogotá
Boston
Calgary
Chicago
Dallas
Detroit
Houston
Mexico City
New York
Palo Alto
San Francisco
São Paulo
Toronto
Washington, D.C.
Asia Pacific
Bangkok
Beijing
Hong Kong
Jakarta
Kuala Lumpur
Melbourne
Mumbai
New Delhi
Seoul
Shanghai
Singapore
Sydney
Taipei
Tokyo
Europe
Amsterdam
Berlin
Brussels
Bucharest
Budapest
Copenhagen
Düsseldorf
Frankfurt
Helsinki
Istanbul
Kiev
Lisbon
Ljubljana
London
Madrid
Milan
Moscow
Munich
Oslo
Paris
Prague
Rome
Stockholm
Stuttgart
Vienna
Warsaw
Zurich
Middle East
and Africa
Abu Dhabi
Doha
Dubai
Johannesburg
Manama
Riyadh
For more information, permission to reprint or translate this work, and all other
correspondence, please email: [email protected].
The signature of our namesake and founder, Andrew Thomas Kearney, on the cover
of this document represents our pledge to live the values he instilled in our firm and
uphold his commitment to ensuring “essential rightness” in all that we do.
A.T. Kearney Korea LLC is a separate and independent legal entity operating under the A.T. Kearney name in Korea.
A.T. Kearney operates in India as A.T. Kearney Limited (Branch Office), a branch office of A.T. Kearney Limited,
a company organized under the laws of England and Wales.
© 2016, A.T. Kearney, Inc. All rights reserved.

Documentos relacionados